INDEX

Explanations

sentences or phrases emphasizing a plot twist or significant event

oai_token-act-pair · gpt-3.5-turbo

periods at the end of sentences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pocket

-0.82

uly

-0.79

 defe

-0.78

 employer

-0.75

oshenko

-0.75

 prepaid

-0.74

cipled

-0.74

 enrol

-0.73

 workforce

-0.72

 scourge

-0.72

POSITIVE LOGITS

 Characters

1.08

 Though

1.04

 Whereas

1.04

 Thankfully

1.03

 Unfortunately

1.02

 Their

1.02

 Especially

1.02

 Throughout

1.01

 Needless

1.01

 Along

1.00

Activations Density 0.883%

sentences or phrases emphasizing a plot twist or significant event

periods at the end of sentences

No Comments

No Known Activations